If you've read this far in the series, you've seen something unusual. You've seen quantum mechanics and Genesis describing the same event. You've seen the Trinity derived from the Born Rule. You've seen the Fall as a phase transition, time as grace, the Cross as entropy reversal. And every piece of it was built on the assumption that Jesus of Nazareth was a real person who really lived, really died, and really rose. So what happens when someone says he wasn't?
The strongest version of that argument comes from Richard Carrier, a historian who applied Bayes' Theorem — the same probabilistic logic used in medical diagnosis, criminal forensics, and signal processing — to the question of whether Jesus existed. His conclusion: at best, a 32% chance Jesus was historical. His book On the Historicity of Jesus is peer-reviewed and methodologically serious. It is not a crank argument. It deserves a real answer. This article gives one. Not by dismissing his method — Bayes' Theorem is correct. The problem isn't his math. It's his reference class.
How It WorksThe Setup
The Bayesian equation is simple: posterior odds equal prior odds times the likelihood ratio. You start with how likely something is before examining any evidence, then update based on what the evidence says.
Carrier's prior probability comes from asking: in the set of all ancient figures most similar to Jesus, how often do they turn out to be historical rather than mythical? He places Jesus in the "heavily mythologized savior heroes" class: Osiris, Hercules, Romulus, Moses, Aesop. Figures who were worshiped, whose stories were filled with supernatural events, and who appeared primarily in sacred literature rather than mundane political records. In that class, at most 1 in 3 turned out to be historical. So he starts with a 33% prior — 1 to 2 odds that Jesus existed.
He examines four categories of evidence: extra-biblical sources, Acts, the Gospels, and the Epistles. He finds most evidence is equally likely whether Jesus existed or not. Some slightly favors mythicism. Three ambiguous passages in Paul slightly favor historicity. The math shakes out to roughly 32% at the upper bound.
The argument is clean. The math is correct given his inputs. The problem is upstream of the math.
The Core ErrorPattern as Fabrication vs. Pattern as Signal
Carrier's entire prior depends on one assumption: the structural similarities between Jesus and other savior-hero figures — incarnation, death, resurrection, salvation, worshiped by followers — are evidence of fabrication. He looks at the pattern and concludes: this is what cultures invent. Jesus shares the pattern. Therefore Jesus was probably invented.
This assumption does real work in his argument. It's the reason Jesus lands in the low-frequency reference class. It's the reason the prior is 33% instead of 75%. Everything downstream inherits it.
What if the assumption is wrong? What if the reason the death-resurrection-salvation pattern appears across multiple cultures isn't that humans keep inventing it — but that the pattern is real?
The Framework's AnswerSignal, Not Noise
This series has argued — with 10 property-by-property confirmations across physics and theology — that physical laws carry the structural signature of the being who made them. We call this the Projection Principle. The Logos projects its nature onto the physical substrate: rationality → mathematics works; self-consistency → universal laws; conservation → Noether's theorem; triadic structure → the Born Rule requires three irreducible operations.
If the Projection Principle is true, then the death-resurrection-salvation arc is not a mythological template that cultures copy from each other. It is the physics of coherence restoration — what it looks like when the entropy drain is overcome by a grace input. It is the shape of reality reasserting itself after a symmetry-breaking event.
Suppose you find that 50 different ancient cultures described a bright light appearing in the east every morning. A Carrier-style argument would say: "The 'sunrise' pattern appears too frequently across too many mythologized accounts. Clearly this is a fabricated motif." The problem is obvious. The sunrise pattern appears everywhere because the sun actually rises.
The death-resurrection-salvation arc appears across cultures because the Logos projects it. Every culture that listened closely enough picked up the signal. Some garbled it. Some mythologized it. But the signal was always real. The question isn't "what percentage of these accounts are fabricated?" The question is: which account, if any, is the source event?
If we grant that the pattern is real, then the frequency of that pattern appearing in mythology does not reduce the prior probability of any individual instance being historical. To the contrary — it raises the question every signal-processing engineer asks: given that the signal is real, which instance is the source?
The CorrectionRunning the Numbers
Correcting only the prior — from Carrier's 1/2 to a more defensible 2/1 based on the correct reference class — and accepting everything else Carrier argues:
Prior 1/2 · Reference class: mythologized savior heroes · Pattern treated as fabrication
Prior corrected to 1/1 · Everything else as Carrier argued · Pattern treated as signal
Prior 2/1 + corrected extra-biblical scoring + framework mechanism for Paul's silences
The entire edifice rests on the reference class. Correct the reference class and the math corrects itself. Even the most conservative correction — moving the prior from 1/2 to 1/1 and touching nothing else — nearly flips Carrier's result to equipoise. The 97% figure requires accepting all three corrections simultaneously, each at the values we chose. A more honest range: 49–97%. But even the floor is devastating to Carrier's conclusion.